Thursday, May 24, 2018

Games as Research symposium, after-action report

A month ago I attended a one-day Games As Research symposium, hosted by TAG at Concordia University and organized by Rilla Khaled and Pippin Barr. If you want my rawest thoughts, here's my live tweet thread from that day.

I learned a lot about design history and current methodologies for studying how a game is made. Here's some of the common topics and threads that we kept coming back to, and a brief summary of each presentation:

What should this research do? What are different ways of thinking about thinking about making games?

Rilla Khaled began the day by arguing that industry devs freely share their processes at events like GDC, but academics can't just study that -- North American academic norms dictate that research must resemble science, which means collecting quantitative data, no matter how pointless it might be. Research should mean more than just numbers, it should also include any kind of investigation, and questioning / defining paradigms should count as research too.

And even if you do like all the number-collection paradigms of research, Jonathan Lessard lamented that such "research" is often based on unquestioned assumptions about making games. Like if you crunch some data about proc-gen platformer levels and machine-learning physics that's great, but there's also a vital unspoken question in that case: why did you choose to work in the platformer genre and what was your process for designing that platformer? How does that game design affect the data and conclusions you're making? It seems pretty problematic to pick any one genre trope as the "pure" generalized medium for games research.

Laureline Chiapello argued all this anxiety parallels general design history, where there was a "crisis of relevance" and how to transfer values from applied science to applied arts, and eventually the somewhat effective solution focused on Donald Schön's notion of "reflective practice." We now imagine design as a "conversation with materials."

Game design as craft. Game design is "design" because there is an intended function for a user, and it is also a "craft" that focuses on the forming of an object (a game) from materials.

Gillian Smith brought a crucial focus to this notion of craft, in her game design experiments with diverse workflows like sewing and egg dyeing. Compared to other types of non-digital craft, game design ends up looking soulless and empty. Sewing a quilt for a family member, for example, symbolizes a relationship and leaves an heirloom to pass on, and all that context can literally keep you warm. Why can't you wrap a video game around you to keep you warm?

Craftspeople also develop a unique relationship with the tools they use. Sewing experts prefer certain sewing machines, and can any game engine truly compare to a well-worn hammer passed through generations of craftspeople? We pay attention to player psychology and flow -- but not to designer psychology, or the phenomenology of game development. A game developer moves their mouse in a specific way, a 3D artist understands a 3D interface differently than someone with no 3D experience. This direction begins moving into usability / HCI research territory, but maybe those fields don't have enough concern for critical theory and aesthetics to be so useful to us as game designers.

New methods and case studies. So we know some of the problems researching games / making games as research, but now let's go into some examples about putting this into practice.

Miguel Sicart proposed a shift from game into play design. When you make that shift and step back from tweaking numbers in a spreadsheet, you can pay more attention to how we use those numbers instead of what those numbers are. Miguel points to Phoebe Sengers' research on "critical technical practice" as a good way to teach design and make good work, with a focus on bringing marginalized concepts to the center of your design.

Pippin Barr proposed a study of version control repositories. If making a game can be an act of research, a Git repo serves as a useful archiving tool to centralize all your thinking and process throughout development. So when you commit your code, you should also commit your mindset, and that way you can track your mindset as you work on your game. Pippin's actually done this a few times now, check out his GitHub examples for "It's As If You Were Doing Work" and "b r 1". If you want to know more about his approach, read about their curiously named "MDMA" (Method for Design Materialization and Analysis) theory.

Stéphanie Mader talked about "Le Village aux Oiseaux" (The Town of Birds), a therapeutic first person photography game to help seniors with Alzheimer's develop their perception and memory skills. Even with a funded dev team and dedicated researchers, it took a long time for them to develop a working relationship with doctors. They had to figure out how to design for Alzheimer's / study their audience, and then they had to formalize their game design clearly enough to communicate it to their subject experts. How much are they asking their players to remember and track? In the end, all that data collection left them with "no time for game design", which suggests we're just scratching the surface of stuff like games and learning / games for change, and we're going to need a lot more rigor and research to figure out how to do any of it properly.

Emma Westecott applies feminist performance theory to understanding games, focusing on the player / designer's cultural position in society. She even ran a feminist game jam called "SYSTEM.FAILURE" where participants were fed meals (!) went home at 6pm (!!) and were paid for their work (!?!). Wow, labor-aware intersectional feminism sounds pretty great huh? But it also suggests an interesting direction for investigating game jams as research.

Owen Gottlieb talked about making historical AR games for kids, and the difficulty of designing and running any type of research experiment involving human subjects in an academic setting. His project had players roleplay as reporters interviewing different historical figures surrounding the notorious Triangle Shirtwaist Factory fire of 1911. He emphasized the importance of documentation for research; they know the game worked only because they got permission to record all the kids' reactions as they played. Owen also pointed to designer Ira Fay's visual documentation for a game called The Tomes (pictured above) where they kept meticulous track of all the different UI screen designs and iterations throughout development.

Oh, and I gave a presentation too, I guess. I talked about how I think of my gay sex games as a sort of "sex design" research, but also a very formal inquiry into writing as a form of game design. That is, I'm one of the few people in games who writes artist statements about my intent and influences -- and I start writing my artist statements early in my development process, to try to check my work and figure out if my thing is actually doing what I say it's doing. Sometimes, I'll end up writing about what I wish the game did, and so I change the game to fit my statement's claims. Then I showed a bunch of sexy hunks to the academics, and much fun was had by all.

* * *

So how does one research games / understand games as research? It seems like a lot of the burden is on designers to develop critical consciousness and foresight to document everything they're doing.

Like as much as I love Pippin's use of the Git repo as a general design archive, I also don't trust any non-developer to understand what's going on in my repo. Any hypothetical non-expert scholars would probably need some kind of code archaeology tool to visualize certain aspects of the repo, or to flag certain commits or code fragments for inquiry / explanation from subject experts or the author. Maybe someone should make this "accessible code review" tool someday?...

After the presentations, we also discussed how to fit this work into the existing machinery of academic research / what a government funding body would want to hear. The idea of research-creation or research-as-creation is still relatively new in academia, which is still reeling from the relative failure or lukewarm reception of the digital humanities. Organizing money and people, doing good work, and then publicizing that work, is still a hard problem.

Still, I left the symposium full of Tim Bits (don't ask), energy, and renewed sense of purpose. At least we have a direction! Now we just have a lot of work ahead of us...